A Route Recommendation System for Sightseeing with Network Optimization and Conditional Probability

This paper proposes a route recommendation system for sightseeing based on a network optimization problem. Traveling and sightseeing times are randomly changed dependent on current traffic and congestion conditions, and hence, Time-Expanded Network (TEN) to contain a copy to the set of nodes in the underlying static network for each discrete time step is introduced. In addition, in order to select the next sightseeing site, conditional probabilities are introduced, which are calculated by current conditions, statistical and Web data. Our proposed model is formulated as a nonlinear and discrete optimization problem, and it is hard to solve it directly and efficiently. Therefore, an efficient algorithm is also developed based on dynamic programming and transformation of the main problem into the recursive equation.